An effective SteinGLM initialization scheme for training multi-layer feedforward sigmoidal neural networks.
Zebin YangHengtao ZhangAgus SudjiantoAijun ZhangPublished in: Neural Networks (2021)
Keyphrases
- multi layer
- error back propagation
- feed forward neural networks
- feed forward
- neural network
- neural nets
- back propagation
- activation function
- hidden layer
- artificial neural networks
- single layer
- recurrent neural networks
- training algorithm
- feedforward neural networks
- neural architecture
- visual cortex
- biologically plausible
- recurrent networks
- genetic algorithm
- learning tasks
- image features
- expert systems
- learning algorithm
- adaptive neural